- cross-posted to:
- [email protected]
- cross-posted to:
- [email protected]
“With fewer visits to Wikipedia, fewer volunteers may grow and enrich the content, and fewer individual donors may support this work.”
Archived version: https://archive.is/20251017020527/https://www.404media.co/wikipedia-says-ai-is-causing-a-dangerous-decline-in-human-visitors/
You must log in or # to comment.
Personally, I have been moving the opposite way. There are so many bullshit websites, wading through them is a pain. Instead, I directly jump to Wikipedia
Same! I have Wikipedia pinned as the first search result on Kagi if there’s an entry
I was thinking of sharing a kagi duo plan with my dad, is it worth it? I currently use ecosia
I have been using it for about a year. There is zero advertisement anywhere. No sponsored posts. Just pure search. And they give you quite some good tools to derank or delist websites for you to define your own quality of search results, it’s very effective. They also go the extra mile of de-ranking websites with many trackers as that apparently correlates with SEO which normally doesn’t track with result quality. It’s a very refreshing take on the web. I love it :)
Edit: they also have an extension which allows you to search while being logged out using tokens which improves privacy, in case you want to go the extra mile.
I have moved the opposite way. I am noticing so much propaganda on Wikipedia that I would rather avoid it.
For example, this page, which is clearly written by Zionist bots, but somehow has edit protection.
AI creating more poor thinkers is how I read this. Its like that outer limits episode where everyone is hooked up to all the answers mentally and one guy can’t have the implant. In the end he is the only who can find his own answers.
I’ve had this thought about heaven before. Like I assume you don’t go in with whatever limitations there are on your brain in your earthly form, which means the people that died longest ago will have this almost unapproachable level of knowledge, and it feels like that everyone would get to that level across eternity. Then what? It would be a space filled with everyone that knows everything. Would you even be you anymore? Wouldn’t it all be boring?
(I don’t believe in heaven)
Like almost every religious concept heaven falls apart almost immediately with even the slightest bit of scrutiny.
Well, you won’t ever be losing me, WP.
We got WordPress’s number one fan here.
I’ll be honest with you: I have absolutely no idea how to interpret that in this context.
WP interpreted as “Winnie the Pooh” instead of “Wikipedia”?
Vanilla guess: WP = Winnie the Pooh.
Ah. I see.
Neither me
More proof that AI ruins everything - if it was needed.
I cancelled my donations to wikipedia since the WMF announced they’d be using AI, and they gave me a really milquetoast bullshit response when I emailed them. Can’t trust such a precious, collaborative, human effort as Wikipedia to a bunch of anti-human robofuckers. 🤷♂️
Correct me if im wrong, but haven’t they since gone back on AI integration, and are no longer planning any, after backlash from their contributors?
Honestly, I don’t know. I stopped listening after they said they were going to.
Too little too late for me really; unless they replace the people who made that decision in the first place, that sort of anti-human, lazy, irresponsible sentiment will remain present in their leadership.
Why do you want to replace people that are capable of changing their minds giving input from other people?
I would say we need more of that kind of leadership.
I don’t know why backing down after massive backlash is being lauded if I’m being completely honest. I want to replace them because the decision in the first place to pollute Wikipedia at all with AI is so poor that I no longer trust them to make good choices.
I was following along when they initially announced it and while I personally think its an awful idea, it did seem to stem not from a profit motive but an intention to make Wikipedia easier to use and more effective, and when the vast majority of contributors and editors responded to tell Wikipedia that it was an atrocious idea and that they shouldn’t do it they listened and scrapped the plan.
My point being, isn’t it better as a whole that they’re willing to consider new things but will also listen to the feedback they get from their users and maintainers and choose not to implement ideas that are wildly unpopular?
Writing off all of Wikipedia (the most effective tool for collaborative knowledge collection in human history) just because they announced a well intentioned tool addition and then scrapped the plan when they realized it was unpopular and would likely degrade their platform seems short sighted at best imo.
enrich the content
I like Wikipedia but it’s like Stack Overflow. Your changes will be reverted by the owner who watches his pages all day long.
It’s like the search summary problem, only worse. Before, people lost traffic already cause Google would try to answer the question with a snippet from your site; however, LLMs (or at least ChatGPT) are trained on Wikipedia, because it’s available data. With Reddit being first, Wikipedia being the second largest source of facts they used (at least, according to statista).
Side note, you’re not even supposed to source facts from Reddit or Wikipedia, they’re better for finding other sources. It’s like a game of telephone, at the tertiary source you’re getting even less accurate information. It’s just oh so stupid the direction end users are herded towards.
That’s a problem across the board. Assuming AI does establish itself, all it’s training data dries up and we basically stagnate.
Also, in this weird inbetween phase until it is actually good, we’ve already generated so much bullshit that AI trains on the hallucinations of other AIs.
As some point we will train it on live data similar to how human babies are trained. There’s always more data.
And the AIs themselves can generate data. There have been a few recent news stories about AIs doing novel research, that will only become more prevalent over time.
Though, a big catch is that whatever is generated needs to be verified. The most recent story I’ve seen was the AI proposing the hypothesis of a particular drug increasing antigen presentation, which could turn cold tumors (those the immune system does not attack) into hot tumors (those the immune system does attack). The key news here is that this hypothesis was found to be correct, as an experiment has shown that said drug does have this effect. (link to Google’s press release)
The catch here is that I have not seen any info on how many hypotheses were generated to find this correct hypothesis. It doesn’t have to be perfect: research often causes a hypothesis to be rejected, even if proposed by a person rather than AI. However signal-to-noise is still important for how game changing it will be. Like in this blogpost it can fail to identify a solution at all, or even return incorrect hypotheses. You can’t simply use this data for further training the LLM, as it would only degrade the performance.
There needs to be a verification and filtering first. Wikipedia has played such a role for a very long time, where editors reference sources, and verify the trustworthiness of these sources. If Wikipedia goes under because of this, either due to a lack of funding or due to a lack of editors, a very important source will be lost.
Wikipedia mad it got competition.
Libertarians be like: free market, competition, no state.
Unless it affects them, then its: how unfair that I can’t have a monopoly, I want the state to remove my competitors.
Is the Wikipedia foundation libertarian? And where do you think the AIs are ripping all their data from?
As far as I know they aren’t some nefarious mega corporation.
AFAIK the founder of Wikipedia is a libertarian who believes the free market will solve all problems. He, of course, has to make a bunch of excuses for why Wikipedia itself isn’t being helped by the Free Market^tm , such as “We don’t have real capitalism bro just trust me bro please donate.”
AI scrapers don’t make donations
I use Gemini for my web searches now. But still donate to the WMF. In most cases I just want an answer with the least amount of bullshit. Even using Startpage to get Google results without Google bullshit, it’s annoying. My wife demands sources, and Gemini will provide those as well.
My view of the danger is that people are sourcing facts from AI instead of Wikipedia. Not worried about volunteers, they’re not going anywhere.
‘facts’
Both are not a source. Both can lead you to a real source.
They would be getting the same misinformation.
I cannot read the article, but this seems like a non-issue.
Loads of old Wikipedia pages are essentially complete. Just freeze them.
And didn’t the Wikipedia foundation have years worth of funding already? And wouldn’t fewer visitors imply less need for server and bandwidth?
Current events need editors, and those will have controversy. I expect primary news sources would be better for anything less than a week old.
A Wikipedia that freezes at 2024 would still be of great value.
Clearly, I am missing the problem.
Old events are not frozen. There are these things called historians and archeologists who are, to this very day believe it or not, still researching “old” events and updating the facts as they find new sources or correct old ones.
Just make wikipedia full of AI content. (/s)
Pretty telling that AI will use most any shitty source besides one that is actually a pretty decent summary of sources.
I’ve been saying this for some time now: AI is going to kill so many business models because it’s really great at creating summaries.
You don’t even need a huge cloud-basrd AI! Local AI—running on your PC—can search the web, summarize the news (and Wikipedia articles), and perform similar tasks without a human ever visiting the site.
It’s like having your own personal secretary that you can tell to go do stuff.
I think it’s going to kill free search engines because it can go do a search on all of them at once in seconds and no human will ever see those ads.
I think it’s going to kill free search engines because it can go do a search on all of them at once in seconds and no human will ever see those ads.
So sites will move to paywalls or aggresive WAFs and those LLMs will become useless at the same time succeeding in killing search engines leaving no viable alternatives.
Already half the public internet is inaccessible without allowing to be heavily fingerprinted and be tracked just to prove you are a human user.
If you think AI is going to stay without ads/subscriptions for much longer you’re not paying attention
Don’t need subscriptions for a local AI friend
Who’s giving you the model?
FOSS nerds, god bless them
